Theoretical and Experimental Research on Voting and Bargaining Games

California Institute Of Technology, Pasadena CA

Investigators

Abstract

This award is for research on several fundamental open questions about interactive strategic behavior in political economy. The approach combines traditional economic modeling, mathematical tools of game theory, and laboratory experiments to study voting behavior and bargaining behavior. In the last few decades, research by theoretical and experimental economists has made tremendous progress in understanding the basic equilibrium forces that underlie behavior in very complex interactive environments. On the theoretical side, detailed mathematical models have been developed to help understand and explain the diverse and subtle factors that affect voter turnout, political polarization, committee decision making, multilateral bargaining, and related phenomena in the realm of political economy. By augmenting traditional economic analysis with the formal theoretical framework of game theory, many new and important insights have emerged. These insights not only help us understand these very complex phenomena, but also provide guidance for how to improve bargaining protocols, voting methods, and collective decision making institutions. On the experimental side, controlled laboratory environments with human subject participants are created as realistic parallels of the situations being modeled theoretically. This allows sharp tests of the hypotheses generated by the theory, and at the same time provides detailed micro-data to enable the measurement of key driving behavioral variables such as risk aversion and social preferences. This in turn feeds back to the theoretical modeling in order to fine-tune the theories to more closely match observed behavior. The specific research to be undertaken encompasses three research areas: (1) The Dynamic Political Economy of Public Goods; (2) Vote Trading; and (3) Income Taxation, Redistribution, and Economic Efficiency. The research in the first area is based on repeated bargaining models under the shadow of a voting rule. Outcomes of the bargaining process evolve over time, with current outcomes affecting opportunities for future bargaining agreements. We study the dynamics of public investment and the effects of voting rules and political institutions on economic efficiency. The second area of study investigates different voting mechanisms that produce outcomes sensitive to preference intensity. Specifically for this study, we are investigating the theoretical and behavioral properties of vote trading and logrolling across issues. We aim to characterize stable vote allocations, focusing on existence, efficiency, and comparison to the majority preference relation. The third part of the project develops equilibrium models to compare the effects of different political institutions on income taxation, redistribution and economic efficiency. The hypotheses generated from these theoretical models are then investigated experimentally.